Ben Williams is AHDB knowledge transfer senior manager. Prior to joining AHDB he ran and worked on a variety of pig and mixed livestock units, from day-to- day husbandry to site management, butchery and further processing
To misquote the great philosopher Biggie Smalls – does more data actually mean more problems? Or is there something to be said about the avalanche of available information out there?
As of 2017, human digital data equated to 2.7 zettabytes or 341 billion three-minute songs. To stack it all into iPads would amount to four structures as big as the Empire State Building. Seems a lot, but if we gave it physical mass it would only weigh roughly 1.6 tonnes, or 0.000002% of the total mass of pork produced in the UK annually.
So, yes, there is a lot of information but we’re not really even scratching the surface of what’s possible. I suspect the problem isn’t the volume of data, but the ability to identify what’s relevant and to present that in a way we can use and extract value from it. In my opinion, these are the challenges facing precision livestock farming and at AHDB we’re starting to look into how to solve a problem like data.
The first issue is joining them up. We know data exits from sire to slaughter. Currently, to join those data points together is a colossal task, involving mountains of data from multiple sources and formats.
But if we increase the amount of data held by each animal, I suspect a link can be made, be it using tried-and-tested technology, such as electronic identification, or emerging technologies, such as cameras, for individual recognition.
Once we know which pig came from which genetic lines, which production processes it went through and what it yielded in terms of product and the associated welfare standards, we can join the dots.
“We don’t have time for this,” I hear people yell. Well, producers shouldn’t have to find extra time as the data and the power to analyse it already exists. So rather than looking at average performance of our herds at distinct points, we should work with companies that have the analytical skills to identify our ‘outliers’. These are the sows, pigs, genetic lines, buildings, processes and individual animals that did not deliver, despite our best efforts. Once these are identified, producers can focus their efforts more effectively – data is starting to work.
Imagine data that tells you which pen always fails you in terms of pleurisy or identifies which sire lines don’t work for you versus which do on a week-by-week basis, or data powerful enough to predict the final slaughter weight of pigs 20 weeks before they are due to be sent to the processor.
All of these are possible and the next step is to see if they lead to businesses that are more profitable.
I hope we are about to see more data to solve our biggest problems. We’ll be talking a lot more about this at Pigs Tomorrow in May.